Pedro Mota

Studying the dynamics of deposits is important for three reasons: first, it serves as an important component of liquidity stress testing; second, it is crucial to asset-liability management exercises and the allocation between liquid and illiquid assets; third, it is the support for a liquidity at risk (LaR) methodology.

Current models are based on AR(1) processes that often underestimate liquidity risk. Thus a bank relying on those models may face failure in an event of crisis. We propose a novel approach for modeling deposits, using panel data and a momentum term.

The model enables the simulation of a variety of deposit trajectories, including episodes of financial distress, showing much higher drawdowns and realistic liquidity at risk estimates, as well as density plots that present a wide range of possible values, corresponding to booms and financial crises.

Therefore, this methodology is more suitable for liquidity management at banks, as well as for conducting liquidity stress tests.

When (Xt)t≥0 is an ergodic process, the density function of Xt converges to some invariant density as t →∞. We will compute and study some asymptotic properties of pseudo moments estimators obtained from this invariant density, for a specific class of ergodic processes. In this class of processes we can find the Cox-Ingersoll & Ross or Dixit & Pindyck processes, among others. A comparative study of the proposed estimators with the usual estimators obtained from discrete approximations of the likelihood function will be carried out.